When should we actually start speech therapy? The research on early intervention timing

Published by Unseen Progress, an independent publisher of caregiver research. Last reviewed 2026-05-10. Part of the speech and language research overview.

Short answer. When concern emerges, not when delay is "confirmed" by another year of waiting. The research is consistent across systematic reviews: earlier intervention produces better outcomes than later intervention, and the cost of evaluating early is small (Law, Garrett, & Nye, 2003; Roberts & Kaiser, 2011). The American Speech-Language-Hearing Association (ASHA) explicitly advises against the "wait and see" default that persists in pediatric practice.

The "wait and see" history

Until the late 1990s, "wait and see" was reasonable advice. Early intervention services were uneven, evidence for their effectiveness was thin, and many late talkers did catch up on their own. Pediatricians were trained to reassure parents, refer at age three if delay persisted, and avoid pathologising children who would resolve naturally.

That advice no longer fits the evidence base. Two things changed:

1. Effective early intervention became widely available. IDEA Part C in the US guarantees evaluation and services from birth to age three at no cost to the family. 2. The evidence for parent-mediated and clinician-delivered intervention strengthened. Roberts and Kaiser's 2011 meta-analysis showed that trained-parent gains are comparable to clinician-delivered gains, and that earlier engagement produced larger effects.

Pediatric advice has not fully caught up with this shift. "Wait and see" is still common in clinical practice, particularly with first-time parents, articulate children, or families perceived as anxious. The default needs updating.

What the research says about timing

Several findings converge on the same recommendation:

  • The Cochrane systematic review of speech and language therapy interventions (Law, Garrett, & Nye, 2003) found that intervention produces measurable gains in expressive phonology and vocabulary, with stronger effects when intervention started earlier rather than later.
  • Roberts and Kaiser (2011) found larger effect sizes for parent-mediated intervention when started before age three than after.
  • The CATALISE consensus (Bishop et al., 2017) endorses early identification at 24 months when concern is present, rather than waiting for a developmental "cliff."
  • ASHA's practice guidance recommends evaluation when any of the 24-month milestones are not met: fewer than 50 words, no two-word combinations, or limited comprehension of simple instructions.

The research does not say every late talker needs intervention. It says every concern warrants an evaluation, and that the evaluation is the gating decision — not delay confirmation a year later.

The thresholds that should prompt an evaluation

These are the parent-observable markers most strongly associated with the recommendation to evaluate now:

  • At 12 months: no babbling, no gestures (waving, pointing).
  • At 18 months: fewer than 10 words, does not respond to name, limited gesture use.
  • At 24 months: fewer than 50 words, no two-word combinations, difficulty following simple instructions, very limited consonant inventory.
  • At 30 months: delay has not closed, sentences are still single words, strangers have substantial difficulty understanding.
  • At any age: loss of previously acquired words or gestures, no eye contact, no joint attention, no pretend play.

Loss of acquired skills at any age is a particularly strong signal that warrants prompt evaluation rather than waiting. So is the absence of joint attention or pretend play, which point toward an evaluation broader than speech alone (see speech delay and autism screening).

What an evaluation actually involves and what it costs

A speech-language evaluation in early childhood is non-invasive and play-based. The clinician will assess expressive vocabulary, receptive language, articulation and phonology, social communication, and motor speech. The session lasts roughly 60–90 minutes.

In the US, early intervention (IDEA Part C, ages 0–3) is free regardless of family income — it is a federal entitlement. Parents do not need a pediatrician's referral; they can contact the state's Part C agency directly. After age three, school-based evaluations through the public school system are also free under IDEA Part B. Private SLP evaluations vary by region; many insurance plans cover at least the evaluation.

If the evaluation finds typical development, you have a baseline and reassurance. If it finds delay, you have a starting point. There is no version of this in which evaluating early is a wasted step.

Why "earlier is better" is asymmetric

The asymmetry of the decision matters. Waiting six months has two possible outcomes:

  • The child catches up. You have lost nothing — you would have evaluated, found typical development, and waited anyway.
  • The child does not catch up. You have lost six months in which intervention is more effective than it will be at age three.

Acting now also has two outcomes:

  • The child catches up regardless. The evaluation produced a baseline; nothing else is lost.
  • The child needed intervention. You started it months earlier than the wait-and-see path would have allowed.

Three of four cells favour evaluating now. The fourth is neutral. There is no scenario in which waiting is the better decision.

What the research suggests doing

1. Treat concern, not delay confirmation, as the trigger. If something feels off, that is enough. 2. Use the published thresholds literally. Fewer than 50 words at 24 months, no two-word combinations, missing milestones — request an evaluation. 3. Skip the pediatrician gatekeeper if needed. You can contact early intervention directly in most US states. 4. Treat loss of skills as urgent. Regression is a different signal from slow progress and warrants evaluation within weeks, not months. 5. Do not be talked out of an evaluation. Pediatric reassurance is sometimes correct and sometimes outdated; the evaluation is the gating decision either way.

Related questions

References

  • Law, J., Garrett, Z., & Nye, C. (2003). Speech and language therapy interventions for children with primary speech and language delay or disorder. Cochrane Database of Systematic Reviews.
  • Roberts, M. Y., & Kaiser, A. P. (2011). The effectiveness of parent-implemented language interventions: A meta-analysis. American Journal of Speech-Language Pathology, 20(3), 180–199.
  • Bishop, D. V. M., Snowling, M. J., Thompson, P. A., & Greenhalgh, T. (2017). Phase 2 of CATALISE. Journal of Child Psychology and Psychiatry, 58(10), 1068–1080.
  • Paul, R. (1996). Clinical implications of the natural history of slow expressive language development. American Journal of Speech-Language Pathology, 5(2), 5–21.
  • American Speech-Language-Hearing Association (ASHA), practice guidance on early identification of speech and language delay.

---

Unseen Progress publishes long-form caregiver research and builds research-backed daily trackers for the families covered. See the full speech and language research overview for the complete framework.